Frank Dellaert
@fdellaert

Robotics & Computer Vision Professor at Georgia Tech, and part-time CAIO at Verdant Robotics. Before: stints at KUL, Skydio, Facebook B*8, Google AI.

 San Mateo, CA
cc.gatech.edu/~dellaert
Joined June 2008
220 Photos and videos
My Annual Reviews article on Factor Graphs in Robotics is finally out with a publicly accessible link: annualreviews.org/eprint/85P…
11
97
482
Article

GTSAM 4.3

After a very busy conference submission period, I was finally able to build and release GTSAM 4.3! 😃 🎉 see gtsam.org. Check out the updated website for the improvements, accompanying

1
14
93
6,015
Can Jev specify graphical models for us?
Article

Jev + graphical models: a paradigm shift?

Could Jev + graphical models fundamentally change how we build systems that reason under uncertainty? Unless you've been off the grid this week, you now know that Jev is @typesafeai’s new model for

23
70
630
140,168
One of the issues I discussed with students in class today is the ethical component: the Bayes nets will inherit whatever biases Jed has internalized from its training data.
1
8
709
Frank Dellaert retweeted
Lu Gan@ganlumomo
Sep 20
🎉 Excited to share our recent work on deformable object manipulation, 𝗣𝗵𝘆𝘀𝗖𝗼𝗥𝗲, accepted to 𝗖𝗼𝗥𝗟 𝟮𝟬𝟮𝟲. Congrats to @hcy1n @ShuohanT! 𝗖𝗼𝗱𝗲 & 𝗗𝗮𝘁𝗮: lunarlab-gatech.github.io/Ph… #CoRL2026 #RobotLearning #DeformableObjects

PhysCoRe: Physics-Corrected Residual World Models for Material-Aware Deformable Dynamics

A physics-corrected residual world model that infers per-particle material from visual observations and corrects a differentiable MPM simulator. CoRL 2026.

lunarlab-gatech.github.io
Haocheng Yin@hcy1n
Sep 19
Excited to share our #CoRL2026 work PhysCoRe: a physics-corrected world model for deformable objects! We couple a physics-based simulator with two feed-forward modules: one infers material from vision, the other corrects its internal dynamics. Check: lunarlab-gatech.github.io/Ph…
9
94
8,012
So proud of my daughter and her PlosOne publication on gene expression in corals, when they are under thermal stress. journals.plos.org/plosone/ar…
24
1,820
Article

Foundation Models and Hierarchical 3D Reconstruction

A special shout-out to Harneet Singh Khanuja, who just presented our paper, “Hierarchical Structure-from-Motion Scales Feedforward Reconstruction,” at the Structure-from-Motion in the Age of Deep

2
15
98
14,132
Some very cool work from @VarunGiridhar3 et al in @animesh_garg’s group - @GTrobotics FTW!
Varun Giridhar
@VarunGiridhar3
Aug 25
You fine-tune a robot foundation model on a hard task. It gets 25%. Now what? Introducing Q-Planning, a learning-based harness that lets large black-box robot policies recursively self-improve. On a hard fine-grained task: 25% → 80% in 100 robot attempts (~30 mins), no extra human data. q-planning.github.io/ 1/n 🧵
2
4
17
5,179
GTSAM is going GPU! 🚀 Ruogu Li has built an impressive new experimental CUDA backend for nonlinear factor-graph optimization in GTSAM, just in time for the 4.3 release next week. Read our blog post: gtsam.org/2026/08/20/cuda-ba… #GTSAM #CUDA #Robotics #SLAM #StructureFromMotion #BundleAdjustment #ComputerVision #GPU #OpenSource
Made with AI
7
15
83
4,777
Follow up PR also made SFM much easier to use, see borglab.github.io/gtsam/sfm/
1
3
694
on bleeding-edge develop, for now.
1
577
@leolrg0102
2
433
Article

FAST-Sync !

A new Georgia Institute of Technology × Northeastern University collaboration—now included with GTSAM! Shane Holmes and David Rosen at Northeastern joined Firat Taxpulat, Yiran Luo, and me at Georgia

1
5
27
2,821
Exciting to see how the community puts @gtsam4 to work! In a guest post, Jash Shah explores an application of factor graphs to 3D Gaussian Splatting, bringing neural rendering together with pose graphs, loop closures, robust noise models, and iSAM2. gtsam.org/2026/08/04/gaussia… #GTSAM #SLAM #3DGS
5
32
3,697
Article

Bringing Geometric Foundation Models to SLAM

Dominic Maggio (@DominicMaggio1), Hyungtae Lim, and Luca Carlone from @lucacarlone1's lab at MIT are pushing an exciting direction for SLAM: bringing geometric foundation models into robotics through

1
15
160
28,249
Yep, that’s me in Paris rn. Same temps as ATL, tbh, but no AC in the AirBnB!
Colin McCarthy
@US_Stormwatch
Jun 22
One of the worst heatwaves in European history is underway. Peak high temperatures forecast this week: France: 45°C / 113°F Monday-Tuesday London: 39°C / 102°F Amsterdam: 34°C / 93°F Berlin: 38°C / 100°F Paris: 41°C / 106°F
4
7
3,484
Frank Dellaert retweeted
Radiance Fields
@RadianceFields
Jun 18
For the people asking about 3D printing gaussian splats, this is honestly amazing.
Dany Bittel
69
281
2,739
193,305
Huzzah! A new GTSAM blog post by Kosuke Inoue: RTK GNSS double-difference factors for pseudorange + carrier phase, with lever-arm variants for GNSS-IMU fusion and example results on a Tokyo urban driving dataset. A great community contribution to GTSAM’s navigation module. Link in replies.
1
8
58
3,947
gtsam.org/2026/06/10/rtk-gns…

Toward Centimeter-Level Positioning with GTSAM: Double-Difference Factors for RTK GNSS

Author: Kosuke Inoue, independent researcher

gtsam.org
1
6
976
This paper is breathtaking! A hybrid of learning and multiview geometry with incredible motion capture results, multi-person *with* contact.
Michael Black
@Michael_J_Black
Jun 5
I’ve been capturing 3D human motion for 30 years and today is maybe the biggest day in that history. We are presenting MAMMA at CVPR (oral session 2A). MAMMA is a markerless multi-camera system that has accuracy similar to marker-based systems.
2
6
80
14,790
Frank Dellaert retweeted
Shangjie Xue@ShangjieXue
Jun 5
Come check out our work on GAVIS, a principled and efficient uncertainty quantification method for 3D Gaussian Splatting active perception. We are presenting at #CVPR2026: 📍 ExHall A, Poster #464, 🕙 10:45–12:45 Come by and chat with us! gatech-rl2.github.io/GAVIS/
2
7
18
1,612
Load more